Incorporating genomics into the cancer clinical trial process.

The effectiveness of current chemotherapeutic approaches for the treatment of solid tumors has reached a near plateau, suggesting we are nearing the limit of cytoreduction. It is hypothesized that this may be due to "subset effect," and that drugs administered according to responses predicted for particular subgroups within the population being treated could overcome what appears to be a limit of cytoreduction. However, the clinical trial process, as currently structured, prevents efficient discovery and validation of predictive markers of treatment response. An alternative process is proposed, based on preoperative therapy and high-throughput multiplexing of markers to provide a built-in, unbiased discovery and validation process for predictive markers. Semin Oncol 28:305-309. Copyright 2001 by W.B. Saunders Company.